## Find a copy online

### Links to this item

## Find a copy in the library

Finding libraries that hold this item...

## Details

Genre/Form: | Electronic books |
---|---|

Additional Physical Format: | Print version: |

Material Type: | Document, Internet resource |

Document Type: | Internet Resource, Computer File |

All Authors / Contributors: |
Aurélien Bellet; Amaury Habrard; Marc Sebban |

ISBN: | 9781627053662 1627053662 |

OCLC Number: | 903883121 |

Description: | 1 online resource (xi, 139 pages) : illustrations. |

Contents: | 1. Introduction -- 1.1 Metric learning in a nutshell -- 1.2 Related topics -- 1.3 Prerequisites and notations -- 1.4 Outline -- 2. Metrics -- 2.1 General definitions -- 2.2 Commonly used metrics -- 2.2.1 Metrics for numerical data -- 2.2.2 Metrics for structured data -- 2.3 Metrics in machine learning and data mining -- 3. Properties of metric learning algorithms -- 4. Linear metric learning -- 4.1 Mahalanobis distance learning -- 4.1.1 Early approaches -- 4.1.2 Regularized approaches -- 4.2 Linear similarity learninG -- 4.3 Large-scale metric learning -- 4.3.1 Large n: online, stochastic and distributed optimization -- 4.3.2 Large d: metric learning in high dimensions -- 4.3.3 Large n and large d -- 5. Nonlinear and local metric learning -- 5.1 Nonlinear methods -- 5.1.1 Kernelization of linear methods -- 5.1.2 Learning nonlinear forms of metrics -- 5.2 Learning multiple local metrics -- 6. Metric learning for special settings -- 6.1 Multi-task and transfer learning -- 6.2 Learning to rank -- 6.3 Semi-supervised learning -- 6.3.1 Classic setting -- 6.3.2 Domain adaptation -- 6.4 Histogram data -- 7. Metric learning for structured data -- 7.1 String edit distance learning -- 7.1.1 Probabilistic methods -- 7.1.2 Gradient descent methods -- 7.2 Tree and graph edit distance learning -- 7.3 Metric learning for time series -- 8. Generalization guarantees for metric learning -- 8.1 Overview of existing work -- 8.2 Consistency bounds for metric learning -- 8.2.1 Definitions -- 8.2.2 Bounds based on uniform stability -- 8.2.3 Bounds based on algorithmic robustness -- 8.3 Guarantees on classification performance -- 8.3.1 Good similarity learning for linear classification -- 8.3.2 Bounds based on Rademacher complexity -- 9. Applications -- 9.1 Computer vision -- 9.2 Bioinformatics -- 9.3 Information retrieval -- 10. Conclusion -- 10.1 Summary -- 10.2 Outlook -- A. Proofs of chapter 8 -- Uniform stability -- Algorithmic robustness -- Similarity-based linear classifiers -- Bibliography -- Authors' biographies. |

Series Title: | Synthesis lectures on artificial intelligence and machine learning, #30. |

Responsibility: | Aurélien Bellet, Amaury Habrard, Marc Sebban. |

### Abstract:

Examines metric learning, a set of techniques to automatically learn similarity and distance functions from data. The authors provide a thorough review of the metric learning literature, covering algorithms, theory, and applications for both numerical and structured data.
Read more...

## Reviews

*User-contributed reviews*

Add a review and share your thoughts with other readers.
Be the first.

Add a review and share your thoughts with other readers.
Be the first.

## Tags

Add tags for "Metric learning".
Be the first.